1. Field of the Invention
This invention relates to signal processing circuits for performing a synchronization acquisition and a frequency (offset) correction in case of using a spread quadrature modulation wave in a spread spectrum communication system. Further, this invention relates to a digital correlator for use in a receiving component of a spread spectrum communication system of the direct sequence modulation type. Moreover, this invention relates to an M phase shift keying (hereunder abbreviated as M-PSK (incidentally, M&gt;4)) modulation/demodulation method employed in a spread spectrum communication system of the direct sequence modulation type. Furthermore, this invention relates to a spreading-code generating method for use in a code division multiple access (hereunder abbreviated as CDMA) system employing a spread spectrum communication method of the direct sequence modulation type.
2. Description of the Related Art
In recent years, it has been studied how a communication network using radio techniques (for example, a cellular network or a local area network (LAN)) is put to practical use. As an example of a communication method employed in such a communication network, a CDMA method using a spectrum spreading has been studied.
Further, two principal types of the spectrum spreading are a direct sequence (hereunder abbreviated as DS) type and a frequency hopping (hereunder abbreviated as FH) type. The DS type of spectrum spreading (hereunder sometimes referred to as DS/CDMA) is a method of effecting communications by directly performing a spectral spreading on an information signal by use of a spreading code or pattern having a frequency which is far higher than the frequency of the information signal (for instance, a frequency which is tens to thousands times the frequency of the information signal). The FH type of spectrum spreading (hereunder sometimes referred to as FH/CDMA) is a method of spreading the spectrum of a narrowband frequency-modulated signal by changing the frequency of a carrier according to a spreading-code pattern and averaging the signal resultingly. Incidentally, a CDMA method is a communication method for performing a multiplexing within a same frequency band by using different spreading-code patterns when effecting a spreading of the DS or FH type.
Previously, a binary phase shifting keys (hereunder abbreviated as BPSK) has been mainly used in case of the DS type of spectrum spreading. Recently, for the purpose of performing data communication at a high speed, a quadrature phase shift keying (hereunder abbreviated as QPSK) by which spectrum spreading and synthesization are performed on each of an in-phase channel (hereunder abbreviated as Ich) and a quadrature (namely, 90.degree.-out-of-phase) channel (hereunder abbreviated as Qch), as well as an M-PSK (incidentally, M&gt;4), has become studied enthusiastically.
In case of employing the spectrum spreading, the influence of what is called a transmission/reception frequency offset of a carrier wave at the time of initial synchronization acquisition is a serious matter. The reason is as follows. Namely, in cases of other communication systems (for instance, an FM broadcasting system and an analog automobile telephone system), a transmission/reception frequency offset can be suppressed by effecting a tracking by use of an automatic frequency control (hereunder abbreviated as AFC) in such a manner to maximize the signal level of a reception signal (namely, a received signal). In contrast, in case of employing the spectrum spreading, the signal level of a desired reception signal is sometimes lower than a noise level until the initial acquisition is completed and thus an AFC as used in the other communication systems cannot be realized in case of a communication system employing the spectrum spreading. Therefore, a method of sweeping the frequency of a carrier within a maximum frequency offset range basically is used in an AFC of the system employing the spectrum spreading. Such a method, however, has a drawback in that it takes time to perform a synchronization acquisition.
Thus a method of calculating the envelopes of components of a reception signal regardless of a frequency offset is employed in the conventional system using BPSK.
FIG. 1 is a schematic block diagram for illustrating the configuration of a conventional synchronization acquisition circuit of BPSK type. In this figure, reference numeral 310 designates a radio antenna; 320 a band-pass filter (hereunder abbreviated as BPF); 330 an AGC circuit; 341 and 342 down-mixers corresponding to Ich and Qch, respectively; 351 a local signal source; 352 a (.pi./2)-phase shifter; 361 and 362 low-pass filters (hereunder abbreviated as LPFs); 371 and 372 analog-to-digital (hereunder referred to simply as A/D) converters: 381 and 382 correlating detectors (hereunder sometimes referred to as correlators) for what is called a de-spreading; 391 and 392 squaring devices; 400 an adder; 410 a synchronization acquisition judgement circuit; 420 a data decoding circuit; 430 a spreading-code generating circuit; and 500 a code clock recovery circuit.
In the synchronization acquisition circuit of FIG. 1, a quadrature detection of a reception signal is first performed to obtain the channels (or components) Ich and Qch. Then, a correlating detection is performed on each of the components by using the same spreading code (namely, the sequence of the same spreading code words). Subsequently, the square of an output of each of the correlating detectors is obtained. In addition, the obtained squares of the outputs of the correlating detectors are added up to obtain the magnitude of the envelope (to be described later) represented on a phase plane. Thereby, the influence of the frequency offset at the time of synchronization acquisition is eliminated.
The above described operation of this synchronization acquisition circuit will be further explained hereinbelow by using expressions or equations (1) to (12). Here, let Dn, C, .omega..sub..smallcircle., .DELTA..omega. and N denote the amplitude of an information signal, a spreading code, a carrier angular frequency, a transmission/reception frequency offset and a reception in-band noise power. Further, the spreading code is C={c.sub..quadrature., c.sub.i . . . c.sub.M-1 } (incidentally, M is a period). Here, a reception spectrum spreading signal y(i) is assumed to be given by the following equation: EQU y(i)=D.sub.n c.sub.i cos (.omega..sub..smallcircle. +.DELTA..omega.)t+D.sub.n c.sub.i sin (.omega..sub..smallcircle. +.DELTA..omega.)t+N (1)
where N is assumed to be given by the following equation: EQU N=N.sub.i cos .omega..sub..smallcircle. t+N.sub.q sin .omega..sub..smallcircle. t (2)
First, a quadrature detection is performed on the signal represented by the equation (1) (namely, the equation (1) is divided by exp(j.omega..sub..smallcircle. t)) to obtain the in-phase channel Ich and the quadrature channel Qch. Then, these channels (or components) are applied to the LPFs, respectively. Thus output signals r(Ich) and r(Qch) of the LPFs corresponding to the input components Ich and Qch are obtained as follows: EQU r(Ich)=D.sub.n c.sub.i cos .DELTA..omega.t+N.sub.i ( 3) EQU r(Qch)=D.sub.n c.sub.i sin .DELTA..omega.t+N.sub.q ( 4)
Subsequently, a correlating detection is performed on each of the signals represented by the equations (3) and (4) by using a spreading code C' which is similar to the spreading code C but may be different from the sequence C only in phase. As is apparent from the definition of the spreading code, an output of each of the correlating detectors can be equal to or greater than a predetermined value only in case where C'=C. If the phase due to the frequency offset can be regarded as constant for one period of the spreading code C', the signal levels corresponding to the channels Ich and Qch are obtained as represented by the following equations (5) and (6), respectively: EQU .SIGMA.{r(Ich).times.c'.sub.i } MD.sub.n cos .DELTA..omega.t(5) EQU .SIGMA.{r(Ich).times.c'.sub.i } MD.sub.n sin .DELTA..omega.t(6)
Then, the squares of the right sides of the equations (5) and (6) are added up as follows: ##EQU1##
As is seen from the equation (7), the frequency offset .DELTA..omega. vanishes and the square of the magnitude MD.sub.n of the envelope is obtained. Thus the judgement on the synchronization acquisition can be effected.
However, in case where QPSK is employed in the conventional system, if the spreading codes C.sub.I and C.sub.Q corresponding to Ich and Qch, respectively, are different from each other and .DELTA..omega. becomes equal to (.pi./2) due to the influence of the frequency offset, a correlation cannot be detected. Namely, the reception signal y(i) in case of this case is assumed to be represented by the following equation: EQU y(i)=D.sub.In c.sub.Ii cos (.omega..sub..smallcircle. +.DELTA..omega.)t+D.sub.In c.sub.Qi sin (.omega..sub..smallcircle. +.DELTA..omega.)t+N (8)
Then, a quadrature detection is performed on this reception signal (namely, the equation (8) is divided by exp(j.omega..sub..smallcircle. t)) to obtain the in-phase channel Ich and the quadrature channel Qch. Subsequently, these channels (or components) are inputted to the LPFS, respectively. Thus output signals r(Ich) and r(Qch) of the LPFs corresponding to the input channels or components Ich and Qch are obtained as follows: EQU r(Ich)=D.sub.In c.sub.Ii cos .DELTA..omega.t-D.sub.Qn c.sub.Qi sin .DELTA..omega.t+N.sub.i ( 9) EQU r(Qch)=D.sub.Qn c.sub.Qi cos .DELTA..omega.t+D.sub.In c.sub.Ii sin .DELTA..omega.t+N.sub.q ( 10)
Here, the equations (9) and (10) can be rewritten as follows by substituting (.pi./2) for .DELTA..omega.. EQU r(Ich)=-D.sub.Qn c.sub.Qi sin .DELTA..omega.t+N.sub.i ( 11) EQU r(Qch)=D.sub.In c.sub.Ii +N.sub.q ( 12)
Thus, before the correlation detection, the component including c.sub.Ii is removed from the signal r(Ich) corresponding to Ich. Further, the component including c.sub.Qi is removed from the signal r(Qch) corresponding to Qch.
Therefore, in this case, the influence of a frequency offset can not be eliminated if the process effected in case of employing BPSK is performed. Consequently, it is difficult to achieve a synchronization acquisition. The present invention is created to resolve such a drawback of the conventional system.
It is, therefore, an object of the present invention to provide a signal processing circuit which can make a synchronization acquisition judgment without being influenced by a frequency offset.
Meanwhile, as the result of the recent progress in LSI techniques or the like, a spread spectrum communication system has come to be applied not only to military or satellite communications but also to industrial or private equipment. Especially, the application of the spread spectrum communication system to a cellular type mobile communications is now studied in the United States and so forth. Thus the spread spectrum communication techniques have rapidly come to draw the attention of the world. Among the various types of the spread spectrum communication methods, the direct sequence modulation method for spreading data by use of spreading-codes referred to as spread signals is advantageous in that a system for performing such a method can be constructed easily by using LSIs and that a distance can be measured by checking a time required to detect a correlation. Thus many research institutes proceed with the development of the direct sequence modulation method.
Hereinafter, a conventional spread spectrum communication system of the direct sequence modulation type will be described briefly.
FIG. 2(a) is a schematic block diagram for illustrating the configuration of a spreading circuit of the conventional spread spectrum communication system. Further, FIG. 2(b) is a schematic block diagram for illustrating the configuration of a de-spreading circuit of the conventional spread spectrum system. In these figures, reference numerals 901, 902 and 909 are 2-input multipliers; 903 and 914 spreading-code generators for generating spreading-codes; 904 and 910 local oscillators for outputting local oscillation signals to be converted by the multipliers 902 and 909 into signals having radio frequencies or intermediate frequencies; 905 a power amplifier for amplifying signals of radio frequencies and transmitting the amplified signals; 906 and 907 a transmitting antenna and a receiving antenna; 908 a receiving front-end circuit for selecting components having frequencies of a required band from received signals of radio frequencies and increasing the levels of the selected components to a necessary signal level; 911 an A/D converter for converting a signal having a frequency of a baseband, which is obtained by the multiplier 909, to a digital signal; 912 a correlator; 913 a synchronizing circuit for monitoring a synchronization acquisition state according to a correlation output signal of the correlator; and 915 a clock generator for generating timing clock signals for the A/D converter 911, the spreading-code generator 914 and so on according to information obtained by the synchronizing circuit 913.
The circuit of FIG. 2(a) is a spreading portion of a transmitting unit. Further, a data signal to be transmitted is inputted to the multiplier 901 from left, as viewed in this figure. Data represented by the data signal is multiplied by a code represented by a spreading-code signal, which is generated in the spreading-code generator 903, by the multiplier 901. Namely, the multiplier 901 outputs a signal, the spectrum of which is spread over the frequencies of the spreading-code signal. Here, pseudo-noise code (PN code) signals are used as the spreading code signals (incidentally, typical examples of a PN code (set) are what is called the M-code and what is called the Gold code). (Incidentally, the autocorrelation characteristics of each of spreading-code patterns and the characteristics of the cross correlation between a spreading-code pattern of a code sequence and another spreading-code pattern of the same code sequence vary with the code.) The spectrum of a data signal to be transmitted is spread by using this spreading-code signal. Further, the spreading of a data signal is sometimes effected at a frequency 2.sup.n -times the frequency of the data signal, for the ease of de-spreading and for the simplicity of the configuration of the circuit.
Next, the signal spread by the multiplier 901 is mixed by the multiplier 902 with a signal sent from the local oscillator 904. Then, a resultant signal is amplified by the power amplifier 905. Thereafter, the amplified signal is transmitted from the antenna 906.
In contrast, in a receiving unit of FIG. 2(b), a de-spreading is effected by performing a reverse procedure of the spreading (namely, performing a demodulation) and the original signal (namely, the data signal) is recovered. First, a signal obtained by increasing the signal levels of a part of a signal received from the antenna 907, which part corresponds to a required band, to a necessary signal level is multiplied by a local oscillation signal of the same frequency as the oscillation frequency of the local oscillator of the transmitting unit, which is issued from the local oscillator 910, in the multiplier 909. Thus a baseband signal, which is spread by using the spreading-code, is obtained. Subsequently, such a signal is converted by the A/D converter 911 into a digital signal. Then, the correlator 912 obtains a correlation value from this digital signal and another signal sent from the spreading-code signal generator 914 which generates the same spreading-code signal as generated in the transmitting unit.
Next, the synchronizing circuit 913 monitors the synchronization acquisition state according to an output of the correlator 912 and controls the spreading-code signal generator 914 and the clock generator 915 to provide a feedback to the correlator 912. Namely, a feedback loop is established in this process. Thus, the circuit 913 operates to stably obtain an output of the correlator and ensure the synchronization.
FIG. 3 is a detail block diagram for illustrating the configuration of the conventional correlator which employs a digital matched filter practically. In this figure, reference numeral 101 designates a shift register for shifting data represented by a signal, which is obtained as a result of the A/D conversion, in response to each sampling clock. This shift register has capacity sufficient to store data sent thereto for what is called a spreading period. Further, reference numeral 102 denotes an arithmetic circuit for producing a product of a spreading-code and data represented by a signal inputted to the shift register 101; and 103 an adder for obtaining a total sum of results of arithmetic operations effected in the arithmetic circuit 102.
In this conventional correlator with the configuration described hereinabove, a signal obtained by performing A/D conversion on the reception signal converted into a baseband signal is inputted to the shift register 101. As shown in this figure, a product of each of signals .gamma..sub..quadrature. to .gamma..sub.(n-1) inputted to the shift register 101 and a corresponding one of spreading-code signals ref.sub..quadrature. to ref.sub.(n-1) is calculated by the arithmetic circuit 102. Further, a correlation output or value is obtained from a result of the addition effected by the adder 103. In case where 1 bit of the transmitted data is synchronized with the spread signal correspondingly to one period of the spreading-code signal, the data can be decoded by making comparisons among the correlation values, for example, by using the maximum and minimum values of the correlation outputs. Further, the frequency of a clock signal can be regulated such that the square of the correlation value maintains its maximum value. Moreover, the same number as of the corresponding spreading-codes or an integral multiple thereof for a clock correction is often selected as the number of stages of shift registers.
The conventional correlator, however, has drawbacks in that a large number of gates such as a shift register, a multiplier and an adder are needed and thus the size of the circuit becomes large and that the power consumption thereof also becomes large. Especially, in case of the conventional correlator of the type that performs a sampling of 1 bit of the spreading-code signal a plurality of times, the operating frequency of the adder 103 becomes high. This drawback affects the realizability of the conventional correlator of such a type. The present invention is accomplished to eliminate the drawbacks of the conventional correlator.
It is, accordingly, an object of the present invention to provide a digital correlator which can reduce the size and power consumption thereof and increase the operating frequency thereof.
FIG. 4(a) is a schematic block diagram for illustrating the transmitting unit of a conventional spread spectrum communication system of the direct sequence type that employs a QPSK. Further, FIG. 4(b) is a schematic block diagram for illustrating the receiving unit of this spread spectrum communication system.
In the transmitting unit of FIG. 4(a), an input signal is first inputted to a QPSK encoder 1101 which performs a mapping of the input signal onto symbol signals representing symbols used in QPSK modulation. Thus data represented by the input signal are converted into two data sequences corresponding to symbols I and Q, respectively. These data sequences I and Q are multiplied by data represented by signals sent from corresponding spreading-code generators 1104.sub.i and 1104.sub.q in modulo-2 multipliers 1103.sub.i and 1103.sub.q, respectively. Then, results of the multiplications are sent therefrom to a quadrature modulation portion 1109 in which PN-code signal is often used as a spreading-code signal. The spectrum of a signal representing data to be transmitted is spread by using the spreading-code signal. In this figure, the quadrature portion 1109 is indicated by being surrounded by dashed lines. Further, in mixers 1105.sub.i and 1105.sub.q of the portion 1109, the respective output signals of the generators 1104.sub.i and 1104.sub.q are mixed with a signal sent from a first local oscillator 1107 and another signal obtained by shifting the phase thereof by (.pi./2). Then, outputs of the mixers 1105.sub.i and 1105.sub.q are added up by an adder 1108. Thereafter, an output signal of the adder 1108 is mixed by a mixer 1110 with a signal sent from a second local oscillator 1111 and thus is converted into a signal of a carrier band. Finally, an output signal of the mixer 1110 is transmitted from an antenna 1112.
In the receiving unit of FIG. 4(b), a signal received from an antenna 1113 is first mixed by a mixer 1114 with another signal sent from a first local oscillator 1115 and is thus converted into a signal of an intermediate frequency band. Then, the converted signal is further converted by mixers 1116.sub.i and 1116.sub.q into baseband signals corresponding to I and Q, respectively, by using a signal sent from a second local oscillator 1119 and another signal obtained by a phase shifter 1117 by shifting the phase thereof by (.pi./2). Such a frequency conversion of a signal to baseband signals by using quadrature signals is called as a quadrature detection. Further, a quadrature detection portion 1118 consists of the mixers 116.sub.i and 1116.sub.q and the phase sifter 1117. Thereafter, correlation values are obtained from data signals (namely, the baseband signals) and spreading-code signals generated by spreading-code generators 1121.sub.i and 1121.sub.q, which are the same as those generated by the spreading-code generators of the transmitting unit. It is, however, rare that the sum of the oscillation frequencies of the first and second local oscillators 1115 and 1119 of the receiving unit is exactly equal to the carrier frequency of the transmitting unit (namely, the sum of the oscillation frequencies of the first and second local oscillators of the transmitting unit). Thus what is called a phenomenon of "(phase) rotation" of data on a phase plane (namely, a phase shift of data) is liable to take place. Therefore, in order to decode data, a correlation is obtained by using 2 correlators 1120a to 1120d. Further, an angle of "phase rotation" (namely, an amount of angular displacement or phase shift) is obtained by a phase detecting circuit 1701. Finally, data is decoded, performing a frequency offset compensation in a QPSK data decoder 1702.
This conventional system, however, has drawbacks in that the configurations of the phase detecting circuit and the QPSK data decoder become complex and that if the transmission rate becomes high, the circuit thereof is limited in processing speed. The present invention is created to eliminate these drawbacks of the conventional system.
It is, therefore, an object of the present invention to provide an M-PSK modulation/demodulation method, by which the configuration of a decoding portion can be simplified without spreading a necessary band in case where a transmission rate is the same as of the conventional system.
It is another object of the present invention to provide an M-PSK modulation/demodulation method, by which good characteristics can be obtained without a frequency-offset compensation circuit.
It is a further object of the present invention to provide an M-PSK modulation/demodulation method, by which characteristics of a spread spectrum communication system can be improved when effecting a multiplexing.
It is still another object of the present invention to provide an M-PSK modulation method, by which the processing speed of a signal processing circuit of a spread spectrum communication system can be increased in comparison with the conventional system employing a QPSK modulation.
Meanwhile, in a conventional system employing a CDMA method, a multiplexing is performed on signals of a same frequency band in order to increase channel capacity. This conventional system, however, has encountered a problem in that such a multiplexing is difficult when using only the M-code or the Gold code.
Further, it is preferable for suppressing the interference between spread waves or signals to use spreading-code patterns having small cross correlation. However, the kinds of the combination of such spreading-code patterns are limited. Moreover, in case of a system of the DS type employing a QPSK (hereunder sometimes referred to as DS/QPSK type), different spreading-code patterns are used corresponding to Ich and Qch, respectively. However, in case where two channels (or components) Ich and Qch are not completely separated in a receiving unit due to a frequency offset after a quadrature detection, if the cross correlation between the spreading-codes respectively corresponding to the channels (or components) Ich and Qch is large, the interference between the components occurs at the time of performing a correlation detection on each of the components Ich and Qch.
As to the problem relating to the channel capacity, what is called the degree of multiplexing (hereunder sometimes referred to as the multiplexing degree) can be increased by lengthening the period of the spreading-code to increase what is called a spreading rate. However, in a practical system, the spreading rate is limited due to the conditions such as the relation between the spreading band width and the information transmission rate and the operating speed of the system. Thus, practically, it is difficult to employ the spreading code of a sufficiently long period.
In case of employing a CDMA method, the relation between the degree of multiplexing and the quality of communication depends on the characteristics of the cross correlation between the code patterns to a large extent. As above stated, the M-code, the Gold code sequence or the like have been studied as the spreading code for a spectrum spreading. For example, it is known that the M-code has good autocorrelation characteristics and thus the peak of the correlation value can be easily detected, though the number of code patterns which can be generated therefrom is small. In contrast, in case of the Gold code, the number of character patterns which can be generated therefrom is larger than that in case of the M-code. However, the Gold code has undesirable characteristics of the cross correlation between code patterns thereof. Thus, when performing a multiplexing, the quality of communication is extremely deteriorated due to the interference between the spread waves or signals. Hence, there is limitation on the number of channels which can be used for communication simultaneously.
Recently, there has been studied a method using an orthogonal code which has good characteristics of the cross correlation between code patterns obtained therefrom. Generally, in case where the orthogonality between code patterns of the orthogonal code is maintained, there is no cross correlation therebetween and thus the degree of multiplexing can be large. In contrast, when the orthogonality is lost, the cross correlation between the code patterns thereof is deteriorated considerably. Therefore, what is called an intersymbol synchronization is necessary in case of employing the orthogonal code in a CDMA system as the spreading code.
Further, in case of employing a Hadamard sequence which is a kind of orthogonal code, a Hadamard matrix having the code length of 2.sup.(n+1) is generated by performing a Hadamard transform on a 2.times.2 Hadamard matrix n times. As can be seen from this generation process, a code pattern consists of repetitive code sub-patterns. Thus, a Hadamard code has undesirable characteristics of the autocorrelation. Consequently, a conventional system employing a Hadamard code has a drawback in that it is difficult to achieve a synchronization acquisition and a multipath separation. The present invention is accomplished to eliminate this drawback of the conventional system.
It is, accordingly, an object of the present invention to provide an M-PSK modulation method by which the interference between the phase components of a data signal can be suppressed.
It is another object of the present invention to provide an M-PSK modulation method by which the degree of multiplexing can be increased.